Page scraping is a frowned up method of working with websites. If you must do it this way, in a small capacity, it may go unnoticed.
To see what is happening with any HTTP traffic, and identify what is happening, you need a HTTP sniffer like:
Fiddler - free web debugging proxy[
^]
Once you have it installed and configured, use the website like you usually would manually and do the upload. Now click on the packet on the left, and look at the request on the right and you will see exactly what is required.